DESCRIPTION: This application focuses on the use of paclitaxel and other chemotherapeutic agents in improving survival in patients with locally advanced (resectable) epithelial cancer of the esophagus (ECE). These investigators have recently found in a Phase I clinical and pharmacologic study that a relatively high dose of paclitaxel (250 mg/m2 over 3 hrs) in combination with cisplatin (CP) and 5-fluorouracil (FU) followed by G-CSF (TCFG regimen) is highly active in advanced ECE and produces a high rate of pathologic complete remission (pCR) in previously untreated, locally advanced resectable ECE. Ten patients with ECE were studied: six had advanced unresectable disease, all had a partial response (two-to-nine months); four had resectable ECE, all had clinical CR, two had pCR. Given the high response of the above trial, the applicants desire to conduct a formal Phase II study of the TCFG regimen as neo-adjuvant to surgery, and to study cellular molecular factors relevant to upstream (e.g., P-glycoprotein) and downstream (e.g., resistance to apoptosis) mechanisms of drug (paclitaxel) resistance in pretreatment ECE tumor specimens obtained from the patients in the trial. They have presented preliminary data obtained from human ECE cell lines that demonstrate that the overexpression of the antiapoptosis Bcl-2 and/or Bcl-XL proteins relative to the proapoptosis Bax protein represent a downstream mechanism of resistance to the cytotoxic effects and apoptosis caused by paclitaxel. They have also observed that loss of cell-cycle G1 checkpoint control represented by the loss of Rb or p16 protein or the overexpression of cyclin D1 and CDK4, as well as by the presence of a mutant p53 protein, are associated with an increased sensitivity of cultured ECE cells to paclitaxel-induced apoptosis. They have also measured, by immunoblot, the expression of Bcl-XL, Bcl-2, Bax, p16 INK4, cyclin D1 and P-glycoprotein in ECE tumor biopsy specimens, and have found that the expression of Bcl-XL, Bcl-2, Bax, and cyclin D1 was highly variable among the tumor samples. Given the above, the specific aims are as follows: 1. To perform a Phase II study of two cycles of TCFG regimen administered pre-operatively to determine the pathologic remission rates, duration of response and survival of patients with resectable ECE. Hypothesis: Addition of paclitaxel, an active agent against ECE that has non-overlapping mechanism of action and side effects with cisplatin and 5-FU would significantly improve the pCR rate in resectable ECE. 2. To determine the expression of Bcl-2, Bcl-XL, Bax and Bcl-XS in the pretreatment tumor samples of ECE, and to correlate these with the quality of remission in the esophagectomy specimen after pre-operative chemotherapy. Hypothesis: low Bcl-2 and/or Bcl-XL to Bax expression ratios in the pretreatment tumor samples correlate with the achievement of a pCR in ECE. 3. To examine the incidence of cyclin D1 and CDK4 overexpression and alterations in the Rb and p16 expressions in the pre-operative ECE tumor samples, in order to determine whether the loss of G1 checkpoint is a potential determinant of sensitivity of ECE to paclitaxel-based chemotherapy. Hypothesis: Loss of G1 checkpoint is common in ECE and is associated with pCR following paclitaxel-based chemotherapy of ECE.